High-performance coated conductors are being developed for various applications such as fusion, NMR, high field magnet, and motor etc. Because the length of coated conductors is in the km scale, completely homogeneous structure over the km scale is not realistic in the actual manufacturing. If the inhomogeneities degrade the properties such as critical current density, the products cannot be shipped. If there is the undetected inhomogeneity in coated conductors, it may disturb the operation of coil and magnet. Various factors affect the coated conductor homogeneity. The control of coated conductor homogeneity should be discussed for the design of nanorod, grain boundary, and process.
For Jc homogeneity, the pinning center structure should be homogeneous. In the pinning study, because the pinning centers are observed at the scale of ~100 nm, the information on inhomogeneity is missing. In this study, we developed the evaluation method of nanoscale pinning center inhomogeneity over the mm scale. Focused ion beam scanning electron microscopy (FIB-SEM) was used to observe the nanorod inhomogeneity at the ~5 mm scale. Based on the result, the influence of the nanorod material on the nanorod homogeneity will be presented.
The Ion beam assisted deposition (IBAD) substrates with the km length scale may contain local orientation degradation, although their average orientation is as small as a few degrees. This means that the grain boundary weak link degrades the local Jc to dominate the overall Jc. To suppress the local grain boundary weak link induced by the substrate inhomogeneity, the Ca doping is effective. Therefore, the Ca-doped YBa2Cu3O7(YBCO) films were deposited on the IBAD substrates to investigate the Jcimprovement by Ca. The Ca doping on the IBAD substrate with Df=5° achieved five times larger Jcthan that in the undoped YBCO, indicating that the Ca doping is effective in achieving the homogeneous Jc to overcome the local weak link. To implement this scenario in the coated conductors, the “find and patch process” based on the Ca doping to improve the grain boundary Jcwill be discussed.
To understand and design the process for homogeneity, the mathematical modeling and machine learning are very effective. The overall Jc is determined by the networked summation of local Jc, and the percolative mathematical modeling is performed. Although the process contains the complicated phenomena that are difficult to describe by the simple equation, the machine learning and mathematical modeling can describe the relationship between the process and the property. The mathematical modeling of percolative nature and the machine learning based process-property modeling will be also discussed for future process design.
This work was partially supported by JSPS, Grant-in-Aid for scientific research 22K18812, CREST, and Research Foundation for the Electrotechnology of Chubu. The IBAD substrates were supplied from Dr. Izumi and Dr. Ibi in AIST.